A host of explanatory risk factors have been linked to the etiology of adolescent alcohol and drug use. In response to the diversity and strength of these factors, prevention programs have targeted a wide range of putative risk factors in an effort to reduce youthful drug use. Perhaps the most instrumental risk factors in terms of prevention interventions are those related to beliefs about the positive and negative consequences of drug use. Most, if not all, prevention programs include strategies to alter beliefs about the expected benefits (anticipated outcomes) of drug use by teaching youth about the adverse physical health, legal, personal, and social problems associated with drug abuse. Despite strong theoretical arguments indicating the use of these strategies, relatively little is known regarding the development of drug-related expectancies, and even less is known about their developmental linkages to other facets of psychosocial functioning. In particular, youth with low self-esteem may affilate with deviant peer groups where direct modeling and vicariously leading help shape positive expectancies for drug ise. In addition to questions regarding their conceptual importance, too few reliable expectancy assessment tools exist, most existing scales have been normed using predominantly white samples, differ in scale construction or response formats, and tap alcohol or tobacco expectancies at the exclusion of other drugs. To address these and related issues, the current study proposes to considerably expand upon ongoing expectancy and etiology research by: (a) psychometrically refining a multi-axial assessment of alcohol and other drug expectancies in a younger cohort (6th grade inner-city, minority youth) than previously examined; (b) prospectively tracking the sample to examine developmental factors influencing expectancies (i.e., conceptual differentiation of expectancies into positive vs.negative); (c) testing the potential mediating mechanisms through which prevention programs are hypothesized to case behavioral change (i.e., drug reduction); and (d) using appropriate statistical and confirmatory modeling techniques to evaluate group differences based on gender, psychosocial risk status, ethnicity, and drug use patterns (i.e., nonuse versus use). The results wull help forge a bridge between etiology and prevention fields as well as furthering our understanding of cognitive motivational factors involved in the etiology of adolescent drug use.